Sweatshops, Opportunity Costs, and

Non-Monetary Compensation: Evidence

from El Salvador1
By DAVID SKARBEK,* EMILY SKARBEK, BRIAN SKARBEK, and
ERIN SKARBEK**
ABSTRACT. Using evidence from field interviews, this article examines
the alternative employment opportunities of thirty-one sweatshop
factory workers in El Salvador and their perceptions about what types
of non-monetary benefits they receive in their current employment.
Interview subjects provide insights into the benefits of their own and
peers employments, their next-best alternative employment, and
other aspects of total compensation. We find that workers perceive
factory employment to provide more desirable compensation along
several margins.

Introduction

To better understand the role of sweatshop factories in developing

countries, this article investigates how sweatshop employment compares to employees alternative employment opportunities and examines the non-monetary aspects of their compensation.2 Economists
argue or assume that people reveal their preferences through their
choices, suggesting that voluntary employment in a sweatshop indicates that it is the workers best available option given those available.3
This article examines the question for a small group of workers in El
Salvador. We identify how favorably workers perceive their alternative
employment opportunities and what role factory jobs play in their
lives. The economic analysis of sweatshop factories usually refers only
to contexts in which workers voluntarily choose their employment.
*Duke University, Department of Political Science, david.skarbek@duke.edu
San Jose State University, Department of Economics, emilyskarbek@gmail.com
Skarbek Law Firm, brianskarbek@gmail.com
**Pierce & Shearer LLP, erin@pierceshearer.com
American Journal of Economics and Sociology, Vol. 71, No. 3 (July, 2012).
2012 American Journal of Economics and Sociology, Inc.

540

The American Journal of Economics and Sociology

Slavery, human trafficking, and employer theft of passports, for

example, are a distinctly different issue and outside of the scope of
this article.4
A workers productivity determines the upper bound on total compensation, and sweatshops often provide capital and technology that
make workers more productive than they would be in other local jobs.
In fact, sweatshop pay often compares favorably with the countrys
standard of living. Powell and Skarbek (2006) examine wage estimates
given by sweatshop critics and find that in nine out of eleven
countries, sweatshop wages equal or exceed the countrys average
income. For seventy-hour workweeks in Cambodia, Haiti, Nicaragua,
and Honduras, sweatshop wages are double the national average
income. In nine out of ten countries for which apparel-related sweatshop wages were available, employees earned wages equal to or
greater than the countrys average national income by working fifty
hour workweeks. Brown, Deardorff, and Stern (2003) document that
multinational firms regularly pay higher wages and provide better
working conditions than local firms. Aitken, Harrison, and Lipsey
(1996) find that higher levels of foreign investment are associated with
higher wages. Lipsey and Sjoholm (2001) find that foreign-owned
firms pay higher wages than local firms do and that wages at local
firms rise when foreign-owned firms are present. Apparel manufacturers and foreign employers often compensate their employees favorably compared to others earnings within their own country.
Workers next best alternative employment opportunities determine
the lower bound on wages. If strong competition for labor exists, there
will be upward pressure on wages.5 Many critics decry large, international corporations opening up third-world factories or relatively
wealthy Westerners purchasing products from these factories.
However, as available alternatives increase in amount and quality,
factory owners will pay their workers higher wages to retain them.
Boycotts may block workers access to the productivity-enhancing
capital and technology that foreign investment brings. In a recent
interesting study, Harrison and Scorse (2010) find that anti-sweatshop
campaigns that led to increases in the Indonesian minimum wage
resulted in large, negative effects . . . on aggregate manufacturing
employment. They attempt to identify the effect of anti-sweatshop

Sweatshops, Opportunity Costs, and Compensation

541

activism on specific districts within Indonesia that produced textiles,

footwear, and apparel (TFA) in the early 1990s based, in part, on who
Nike employed as vendors in 2004. With this data, they find that
while anti-sweatshop activism did not have additional adverse effects
on employment within the TFA sector, it did lead to falling profits,
reduced productivity growth, and plant closures for smaller exporters.6 Basu and Zarghamee (2009) model consumers choices to
boycott products made by child labor and find that the boycott can
actually lead to a rise rather than fall in child labor because of a
backward bending household labor supply curve.
The relevant alternative employment opportunities to sweatshop
labor are often much worse. For example, journalist Nicholas Kristof
(2009) reports on children who survive by picking through a vast
garbage dump for plastic in Phnom Penh, Cambodia. These children,
he reports, hope to someday work in a sweatshop instead of scavenging through the dangerous, dirty trash heaps (see also Kristof
1998). After the introduction of U.S. anti-child labor legislation in
Congress in 1992, Bangladesh textile factories terminated the employment of an estimated fifty thousand illegal child laborers. UNICEF
(Bellamy 1997: 60) reports that the children had little to no access to
education and found employment in stone-crushing, street hustling
and prostitutionall of them more hazardous and exploitative than
garment production. For third-world workers, sweatshops often
provide much better employments than their available alternatives.
While Kristof (2009) provides useful anecdotal evidence, this article
seeks to provide a more systematic examination of the alternative
employment opportunities of sweatshop workers and factory compensation with evidence obtained through open-ended field interviews of thirty-one Salvadoran sweatshop workers.
Heterodox approaches, such as the Marxist and Old Institutionalist
literatures, provide alternative perspectives to mainstream economics
on scarcity, choice, and wage labor. They emphasize historic-empirical
analysis rather than formal modeling, dynamic and conflictual processes rather than static equilibrium, the possibility that economic
forces lead to domination and power relationships, methodological
collectivism, an interest in want-creation, and embrace a valuedirected inquiry (Dugger 1996). Feminist economics shares many of

542

The American Journal of Economics and Sociology

these tenets and emphasizes the importance of unpaid, household

labor and the role of race, gender, and class (Power 2004).
Like these heterodox approaches, our study emphasizes a rich
contextual analysis of social phenomena rather than a focus on formal
modeling, allowing for examination of out-of equilibrium states of the
social provisioning processes. Our study is an attempt to collect,
document, and understand the qualitative, historical accounts of
populations traditionally viewed as marginalized (low income earners
and females). The majority of workers that we interviewed are female
and all of them are low-skilled by developed country standards, and
as such, our study focuses on the context of decisions made by a
subordinated group and examines the interplay between market
exchange and household production activities (Power 2004). In addition, we examine how factory work relates to Amartya Sens (1999)
emphasis on the importance of obtaining capabilities such as education. All of the aforementioned factors typically occupy the forefront
of heterodox studies of labor and development and auxiliary positions
in similar mainstream studies.
Perhaps most clearly consistent with these heterodox approaches,
our research methodology explicitly recognizes that people give
meaning to their economic activities, and these provide a necessary
foundation for understanding the social provisioning process. As
Dugger (1996: 40) explains about the people engaged in the social
provisioning process, [t]hose other people can talk. They can write.
They can tell their own stories. What economists need to do is learn
how to ask questions and learn how to listen to peoples stories. By
adopting an open-ended interview research method, we attempt to
access the meanings that people give to their jobs and make these
central to understanding factory work and labor supply decisions.
Despite the similarities with heterodox literatures, this article
remains primarily within the mainstream economics perspective. We
focus on the traditional neoclassical assumptions of preference revelation through choice and use methodological individualist economic
theory to inform our survey instrument and research design. Central to
our study is the continuance with the assumption that scarcity is a
part of the human condition rather than part and parcel of life within
our advanced capitalist economy because of want-creation by adver-

Sweatshops, Opportunity Costs, and Compensation

543

tisers and corporations (Matthaei 1984, 91). While we recognize that

heterodox theories of wages and labor may not agree with the
concept of scarcity as an ever-present relationship between desirability and availability within real resource constraints, we maintain that
the opportunity cost concept is important and fruitful for understanding sweatshop labor decisions.7
Methodology

This research focuses on El Salvadoran factories because journalists and

social activists have frequently criticized its sweatshops and the
National Labor Committee has actively been involved in workers rights
issues there since the 1980s.8 At an Adidas shareholders meeting in
November 2008, Sonia Lara Campos (2008), an activist with the National
Labor Committee in Central America, criticizes Salvadoran apparel
manufacturers and asks, [h]ow long do the workers of Adidas have to
wait until they receive a dignified wage? Campos (2008) reports that
worker compensation barely provides for the food needs of a family of
four and is insufficient for a broader basket of necessities.
Through an in-depth examination of subjects alternatives to and
compensation in Salvadoran sweatshops, we bring evidence to bear
on these claims as well as provide evidence regarding the role of
sweatshops in the lives of El Salvadorian wage earners. We collected
these data in early 2009 from subjects employed in two factories in El
Salvador: the Partex Apparel Group and Royal Textiles. These factories
produce garments for organizations such as Adidas and the United
States military. Interviewing subjects at two factories is not necessarily
representative of factory work in general or representative of El
Salvadoran factories in particular, but this method provides two advantages. First, to the extent that this sample contradicts critics claims, it
reveals that there is greater heterogeneity in the welfare implications
of factory labor than is commonly implied in popular and public
policy discussions of sweatshops. Second, these interviews provide a
rich description of how factory employment affects a particular group
of subjects in a developing country, so they provide a more accurate
and complete depiction of sweatshop work than relying only on
wage data.9

544

The American Journal of Economics and Sociology

Our study focuses on the subjects perspectives of their environment, employment, and their relevant opportunities. In total, we
conducted and recorded thirty-one interviews with employees.10 The
sample consists of approximately fifty-five percent females, and subjects are, on average, twenty-nine years old. Interview subjects at both
factories work as either cutters of material or machinists who sew the
products. All potential interview subjects work in these same two jobs.
We used both purposive and random sampling methods for selecting
interview subjects; we asked for volunteers and we randomly selected
subjects from alternating project groups based on where they were
sitting in the factory. The only difference in responses between these
two groups is that a greater percentage of the subjects who volunteered rather than being randomly selected reported payment related
problems, thirty-eight percent versus seventeen percent. This may
reflect a greater desire to voice concerns on the part of volunteers than
randomly selected subjects. Recognizing that the views of employers
may also be of interest, we conducted interviews with managers of
each factory. We conducted all interviews in private and with strict
confidentiality. We communicated the confidentiality protocol to the
subjects both verbally and in writing in both English and Spanish. We
conducted the interviews with a local Spanish interpreter and
recorded the entire process. A transcription service in the United States
then retranslated and transcribed the recordings. This ensures that no
miscommunications or incorrect translations might have occurred
between the interviewer, subject, and translator. These measures
provide assurance to interview subjects and help to elicit truthful
responses and ensure accurate communication.
Given that our central question concerns why subjects choose
their current employment, how they value the non-monetary components of compensation, and how they perceive their relevant
alternatives, qualitative research methods offer several advantages.
First, in seeking to understand the non-monetary components of
compensation, we necessarily are dealing with few quantitative variables. A workers assessment of the various non-monetary aspects of
employment in a particular job is multifaceted. It is this diversity and
subjectivity of non-monetary compensation that makes generating
precise measurements of total compensation infeasible. As past

Sweatshops, Opportunity Costs, and Compensation

545

research has focused on quantitative variables, such as wages and

employment (Powell and Skarbek 2006; Harrison and Scorse 2010),
this article hopes to provide a richer (though admittedly still imperfect) depiction of compensation by focusing on the non-monetary
components. Second, the open-ended nature of an interview (as
opposed to a survey instrument) invites respondents to supply their
own interpretive framework to the question and have their perspective reflected in the answer. Survey instruments provide a list of
possible responses, which allows researchers to increase sample
size, but at the expense of restricting the set of responses to a range
selected by the researchers. Smaller sample sizes result from using
open-ended interviews, but this method provides more informative
results. The aim of the research, in fact, was to identify alternatives
that we as researchers do not know. Because of the open-ended
interview process, there are not thirty-one answers to all of the
questions. However, our data are consistent with the standards in
the literature when using this research method. For example, a
recent book (Esbenshade 2004) on sweatshops and monitoring in El
Salvador relies on one hundred thirty nine personal interviews, but
only eighteen of which are with Salvadoran factory workers. Given
that the aim of our research is substantially narrower, we feel that
thirty-one interviews is sufficiently large to provide evidence and is
consistent with the literature.
Our central research question is explicitly concerned with privileging the respondents narratives over the researchers prior understanding of possible responses (Weiss 1994). This technique attempts to
overcome for the imposition of Western standards on the opportunity
sets of workers in developing countries. For example, Bhagwati (2004)
has argued that policymakers often develop advice and standards
based on the tradeoffs faced by Westerners, not by the individuals
living within developing economies (see also Hall and Leeson 2007).
By allowing the interview subjects to frame and inform their responses
with personal experience, the subjects provide greater insight about
what perspectives and beliefs guide their action, and thereby avoid to
some extent the preconceived reference points of the researcher
(Denzau and North 1994) and provide evidence for policymakers
about their actual tradeoffs.

546

The American Journal of Economics and Sociology

Alternative Opportunities for Sweatshop Workers

We designed our interview instrument to elicit responses about what

subjects perceive to be their alternative employment opportunities and
how they value non-monetary aspects of their employment. This
might not always be obvious to an individual, so we asked a variety
of questions about subjects previous employment, pay and working
conditions, peers employment situations, and alternative opportunities. By examining together the answers from these various questions,
the interviews provide a strong indication of how subjects perceive
their employment relative to their alternatives.
To begin, we report on a formal policy that affects compensation:
El Salvadors minimum wage. Table 1 identifies the governmentmandated daily minimum wage in different industries.
Compared to other industries, clothing and manufacturing factory
work has the third highest minimum wage. The subjects in these jobs
benefit from the higher wage relative to workers in other industries.
We asked subjects if pay at their current employment was sufficient
to provide for their needs. Forty-eight percent of respondents
answered in the affirmative with no qualifications (see Table 2).
Forty-one percent answered in the affirmative, but noted hardships in

Table 1
Salvadoran Daily Minimum Wage by Industry
Business and Service Industry
Industry (not clothing and manufacturing)
Clothing and Manufacturing Factories
Seasonal Agriculture/Coffee
Harvesting Coffee
Seasonal Agriculture/Sugar Mills
Agriculture and Livestock Industry
Harvesting Sugar Cane
Harvesting Cotton
Source: El Salvadorian Labor Code, Edition 62a, July 2008.

$6.41
$6.27
$5.57
$4.34
$3.28
$3.16
$3.00
$2.78
$2.50

Sweatshops, Opportunity Costs, and Compensation

547

Table 2
Are you able to manage financially with what you earn?
(n = 27)
Yes, unqualified
Yes, qualified
No
Managed with the assistance of family

13
11
1
2

Source: Interviews.

doing so. One subject answered that her current pay was insufficient
to meet her needs, and two subjects noted that they also rely on the
contributions of family members.
One subject explains that sweatshops provide important nonmonetary compensation:
I do like to work here. I dont like working in the fields anymore. I make
more money here so its better here. There are also other benefits the
company provides and that help us at home and with the family. . . . I think
that working in a factory you have the benefit of having medical insurance,
medicine, and benefits such as now they are providing school packages for
our children who go to school. They give us vouchers for shoes. Those are
benefits you can have at a company. In a job as a housekeeper, you dont
get any of that. It does have its advantages sometimes because you dont
have to pay for transportation or food but you dont get any other benefits.
While here at the factory, you do have some things. You can go to them
for assistance and if they can help you, they do (Subject 10).

Past research that examines only monetary compensation of factory

workers will underestimate the benefits of factory employment by
ignoring these many additional forms of compensation. Another
subject explains, The textile factory work is much more favorable
because many of the stores dont provide insurance, [social security] or
any of the benefits that the law says they should provide. One
practically only has the salary and one even has to figure out a way
to eat breakfast and lunch from that salary. Things like that. The
factory work is very good (Subject 6).
Table 3 presents subjects appraisals about many facets of sweatshop employment, including wages, working conditions, location,

548

The American Journal of Economics and Sociology

Table 3
How Do Your Current Wages and Conditions Compare to
Your Previous Employment? (N = 24)
Specific
Subject

1
3
4
6
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
28
29

General

Wages

Conditions

Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
N/A
N/A
Improved
Improved
Improved
Improved
Improved
N/A
N/A
N/A
N/A
Worse
Improved

+
+

Stability

Location

Benefits

Schedule

+
+
+
+
+
+
+
+

+
+
+

+
+

+
+
+
+

+
+
+
+

+
+

+
+
+
+
+
+

+
+
+

Source: Interviews.

stability, and other non-monetary benefits. Some subjects reported

higher wages but noted that other, less-desirable factors offset this to
some extent. When asked about their current employment relative to
previous employments, one subject reported her current total compensation was worse than in previous jobs (see Table 3). The subject

Sweatshops, Opportunity Costs, and Compensation

549

explains, In some ways yes [it is a better job], in some ways no,
because there I was closer to my family. Here, in contrast, Im far from
my children all day long. So, no. Its not better here (Subject 28).
Table 3 provides a rough summary of workers views about various
aspects of their employment. The subjects indicating specific benefits
of sweatshop employment over previous employment are indicated
by a + and those indicating a cost associated with their current
position compared to previous employment are indicated by -.
Those subjects marked as N/A gave only descriptions of the relative
costs and benefits of current employment without specifically indicating that they thought these attributes made their current employment,
on net, better or worse than previous jobs.
We asked how subjects viewed their jobs relative to their peers jobs
(Table 4). Seven reported that their peers earn more, and seven
reported peers earning less than they do. Three interviewees reported
that their peers had better working conditions, and nine people
reported that their peers had worse working conditions. Four subjects
responded that peers employment or compensation is less certain; for
example, compensation in factory work was explained to be more
certain from week to week relative to street vending or agricultural
work.
Sweatshop critics allege widespread abuse of employees. To
examine this issue, our study asked subjects if employers had mistreated them or if they had ever heard of employers in other locations
mistreating workers. Table 5 presents the subjects assessments of
employer mistreatment. Because employer abuse can take many

Table 4
How would you describe your peers jobs?
Earn More
Earn Less
Better Working Conditions
Worse Working Conditions
Earnings Less Stable
Source: Interviews.

7
7
3
9
4

550

The American Journal of Economics and Sociology

Table 5
Employer Treatment of Subjects (n = 28)
No Conflicts
Verbal Conflicts
Payment Related Conflicts

14
12
8

Source: Interviews.

forms and range in severity from isolated verbal incidents to systematic nonpayment, or even involve physical or sexual assaults, it was
especially important in this case to privilege subjects responses over
the suggestions of researchers.
Fourteen respondents reported no conflict with either former or
current employers. Twelve subjects reported conflict with employers,
giving examples of verbal conflicts. For example, one subject explains,
Sometimes . . . the bosses treat [workers] badly, they pressure them to
increase production, theres a lot of pressure put on them in other
places. They want more and more production. Thats what they say.
Another subject recalls a conflict with an abusive manager by stating
that, . . . more than anything else, its insults when people are
working, to get them to hurry up by offending them. . . . [At this
factory], it was a supervisor, like she was screaming at them and
insulting them. But they solved that quickly, and they fired her [the
supervisor] . . . A subject explains, at [my past job] it was very
difficult because people treated us with very rude words, nicknames,
things like that (Subject 6). Ten subjects reported hearing their peers
talk about workplace conflicts, but did not experience any problems
themselves: Sometimes when you travel by bus and there are many
people going to work you hear them say that the bosses treat them
badly, that they pressure them to increase production. . . . Not where
I have worked, but you hear people say those things when they talk
to their friends (Subject 25).
The failure to pay subjects presents a serious problem because it
violates the foundational agreement of the exchange, which presumably made the voluntary participation of the subject beneficial ex-ante.
Table 6 summarizes all of the payment related incidents found in our
study.

551

Sweatshops, Opportunity Costs, and Compensation

Table 6
Payment Related Incidents Reported by Subjects (n = 8)

Employee

Personally
Experienced or
Attributed to Qualified as
Heard About
Nonpayment or
Closing of
an Isolated
From Others Delayed Payment Sweatshop Occurrence

Subject
Subject
Subject
Subject
Subject
Subject
Subject
Subject

Second Hand
Second Hand
Experienced
Second Hand
Second Hand
Experienced
Experienced
Experienced

1
5
6
7
9
19
20
24

Nonpayment
Nonpayment
Delayed
Delayed
Delayed
Delayed
Nonpayment
Delayed

No
Yes
No
No
Yes
No
Yes
No

Yes
No
No
No
Yes
No
No
Yes

Source: Interviews.

The first three columns show that four subjects reported hearing
about employers paying subjects late or not at all and four personally
experienced such problems in former sweatshop employment (Subjects 6, 19, 20, 24). None of the subjects interviewed reported payment
related problems in their current employment. The third column
shows that five of the eight incidents involved subjects receiving late
payment for their labor services, while three occurrences involved no
remuneration at all for their labor services.
Of the three accounts of nonpayment, two subjects attributed the
failure to pay to the closing of the factory, which was also a reported
reason for delayed payment. For example, Subject 20 recounted, I
worked for [a factory] and they didnt pay me. . . . They closed the
factory and they didnt pay us. Several other subjects identified the
same factory, which had subsequently closed down, as having a bad
reputation for treating workers poorly.
One subject explains that factories may seem bad but they provide
an opportunity to increase ones earnings: Many people say that in
those businesses [factories], you are forced to work, but you are not
forced; you make an effort to make more money. Moreover, when

552

The American Journal of Economics and Sociology

Table 7
Prior Generations Employment
Mothers Employment
Occupation

Homemaker
Agriculture
Vendor / Crafts
Housekeeper
Manufacturing
Construction
Other
Total

Fathers Employment

Responses

Percent

Responses

Percent

8
9
4
2
1
0
1
N = 25

32%
36%
16%
8%
4%
0%
4%
100%

0
15
1
0
3
3
1
N = 23

0%
65%
4%
0%
13%
13%
4%
100%

Source: Interviews.

asked about abuses, some subjects explained that employers treated

them well: For the two years that Ive been working here, I feel that
a situation like that [a problem with the employer] hasnt happened.
On the contrary, they usually try to help us because the majority of us
are women and single mothers. There are always benefits and good
incentives . . . (Subject 9). Another subject, in response to a question
about abuse, said, . . . the bosses are very approachable. They talk to
us. Even outside of working hours, they talk to us and joke with us
(Subject 6).
Economic development often involves more than just higher wages
or improved working conditions. One broad measure of economic
development is to compare the opportunities of sweatshop workers to
the employment options of the previous generation. Table 7 summarizes findings for the employment activities of the subjects mothers
and fathers. Interviews revealed that thirty-two percent of subjects
grew up with their mothers at home. Another sixteen percent of the
subjects mothers earned income by selling crafts or foodstuff produced in the home. Of those women who did work outside the home,
agricultural day labor was the most common occupation.

Sweatshops, Opportunity Costs, and Compensation

553

One subject described her decision between working in the home

and going to work in the factory, indicating that for women entering
the workplace it is an improvement:
They pay me $35.00 per week and I can manage with that. But what if I
wasnt able to make that much working from home? If you work from
home, you also spend money daily, and sometimes, we spend more than
we earn. Sales are not a good business right now. When I have a
conversation with someone, I tell them that we better hope that the
factories wont begin to flag because the factories help people a lot.
Factories and businesses like one this help women a lot (Subject 17).

Agricultural work represented the most common employment of

subjects fathers. Three subjects indicated that their fathers worked in
various types of construction, and two others indicated employment
opportunities of street vending or service related activities. Of the
forty-three parents discussed in the interviews, only four held jobs in
manufacturing and only one of those was held by a woman.
One subject explains, [my] parents always say, you should be more
than I am so you wont have to work so hard. I will provide you with
an education so thatnot that you will make very much, but that at
least you wont have to work so hard, so you wont have to work in
the rain or under the sun (Subject 25). This subject perceives factory
employment as an improvement over the agricultural employment of
former generations.
Amartya Sen stresses the importance of certain types of improved
non-monetary factors, termed capabilities, such as life expectancy,
infant mortality, malnutrition, and the environment as indicators of
economic development. Sen emphasizes the importance of access to
education (1999: 41, 90, 108, 129).
Salvadoran sweatshop workers perceive themselves as receiving
higher wages and better working conditions than in other employment, and these jobs facilitate their childrens access to education.
According to the United Nations (2010), in El Salvador the primary
education dropout rate for all genders is approximately thirty-one
percent. There were twenty-six children among the fourteen interview
subjects with children in our study. Excluding those children who did
not attend school or who could only do so with family assistance,
eighty-eight percent of these children were currently enrolled in

554

The American Journal of Economics and Sociology

Table 8
Can you manage to send your children to school? (n = 14
subjects with a twenty-six school age children)
Yes
Yes, with family assistance
No

23
2
1

Source: Interviews.

school (see Table 8). Family assistance allowed two children to attend
school who would not have been able to do so solely with sweatshop
wages. One subjects child, eighteen years old at the time of the
interview, dropped out of school after the ninth grade. As the
employee reported, My son no longer wanted to go to school but he
is learning to be a mechanic in an auto shop (Subject 10).
After we asked one subject how long she plans to send her child
to school, she responded, Well, I think that while I have a stable job
or if I can find a better one, I think until he goes to university. Until
he becomes independent and can have a life that is good for him,
a life like the one I wasnt able to have (Subject 1). One subject
explains:
With the way the situation is right nownow that I am at this company,
its not that I can do everything; but now that I am here, I feel that I will
be able to [send the children to school]. When I earned less, I only had
enough to buy one school uniform, just a few notebooks; I would buy
them little by little. Now that I am working here and I am making a bit
more money, I feel that I can balance the expenses better (Subject 9).

These subjects believe that the increased earnings from sweatshop

labor facilitate acquisition of greater capabilities through increased
access to education. This is consistent with the findings by Subrick and
Boettke (2002), who find empirically that increased economic development correlates strongly with improved measures of capabilities.
For Salvadoran workers, factory employment can play an important
role in intergenerational mobility and enhanced educational opportunities for their children.

Sweatshops, Opportunity Costs, and Compensation

555

Conclusion

Few people in developed countries would enjoy working in a thirdworld sweatshop factory, as the pay and working conditions are
generally much worse than opportunities in developed countries. The
options available to unskilled labor in the developing world are
context dependent, influenced by the interaction between a variety of
institutional and historical factors. Our study privileges the workers
own perceptions and narratives about what the relevant alternatives
are for their labor. As a result, this method better allows the subjects
(not the researchers) to impart their own understanding about the
nature and scope of their labor supply choices. In contrast to more
opaque quantitative analysis, these narratives more easily convey the
beneficial and perverse outcomes that can result from spontaneous
orders (Martin and Storr 2008; Dugger 1996).
Field interviews reveal that subjects perceive their alternatives,
including agricultural work and street vending, as less desirable when
compared to sweatshop labor. Non-monetary benefits are an important part of this appraisal. The interviews provide information about
the margins along which subjects compensation improves and identify factory employment as one means of improving intergenerational
mobility, educational attainment, and improved economic opportunities for women. This analysis suggests two implications for sweatshoprelated policy and economic development more broadly.
First, policies for developing countries that are based on legal
standards in developed countries can undermine development
because they differ from the actual communitys relevant opportunity
sets.11 Hall and Leeson (2007) investigate the claim that governmentmandated labor standards currently in place in developed countries
are appropriate for developing economies. Finding that the United
States was much wealthier at the time the country adopted different
labor regulations, they conclude that imposing similar labor standards
on developing countries involves an income tradeoff inappropriate for
their current level of wealth. Our research similarly finds that sweatshop labor policy based on Western standards may be inappropriate
because workers in developing countries have fewer opportunities
and are less productive than workers in developed countries are.

556

The American Journal of Economics and Sociology

Focusing on the actual opportunity sets of employees (and potential

employees) in these countries can improve policy by recognizing a
do no harm approach to adopting labor standards. While labor
regulations in the United States may be tolerable, requiring employers
to meet certain wage levels, working conditions, or benefits programs
can do substantial harm to poor workers by eliminating their most
preferred employments. For example, Harrison and Scorse (2010) find
large, negative effects on aggregate manufacturing employment from
the imposition of a higher minimum wage motivated by the U.S.
governments threat to withdraw special tariff privileges because of
human rights issues.
Second, our study of the employment opportunities of sweatshop
workers is relevant to understanding the process of economic development more generally.12 Jeffrey Sachs (2005: 11) argues that factory
manufacturing is an important step in developing a more prosperous
economy, noting, sweatshops are the first rung on the ladder out of
extreme poverty. In fact, we find that sweatshops are several rungs
higher on the ladder. By showing that subjects are indeed choosing
employment opportunities that improve their wellbeing and capabilities, our results are consistent with Sachs arguments about the benefits of sweatshop employment.

Notes
1. The authors thank Chris Coyne, Cynthia Mertens, Matt Zwolinski, two
anonymous referees, and the Editor for helpful comments.
2. Though scholars define sweatshop in varying ways, the term usually
refers to factories in developing countries that produce goods, often apparel,
for multinational corporations and consumers in developed countries and
employ workers for (what the speaker believes are) low wages and poor
working conditions. For a more thorough discussion of the definition of
sweatshops, see Zwolinski (2006). More broadly, Zwolinski (2007) provides
an important discussion on the ethical issues surrounding sweatshops
3. Social scientists in fields other than economics and outside of orthodox
economics have been especially skeptical of such a claim, in part because of
differing perspectives on how people form preferences and the meaning of
opportunity cost itself. Dugger (1996), for example, perceives individuals as
possessing a fetishism of commodities and suggests that people are buying
more and more consumer commodities without ever receiving any intrinsic,

Sweatshops, Opportunity Costs, and Compensation

557

consummating satisfaction from them. Consistent with this logic is continued

employment at a job that provides no intrinsic, consummating satisfaction for
the employee.
4. On this point and a case for sweatshops in the human rights context,
see Powell (2006).
5. In a perfectly competitive labor market, these upper and lower bounds
will simply be conditions that characterize equilibrium.
6. Harrison and Scorse (2010) note that the identification of TFA districts
in this way is an important limitation because these may not be the firms and
locations used in the early 1990s. As a result it is difficult to sign the direction
of the bias, since it is equally likely that only the pro-worker or anti-worker
vendors have survived.
7. While Marxists may reject the concept of opportunity cost, we assert
that all choiceeven choice in constrained environmentsnecessitates tradeoffs on the margin. Our framework views labor decisions as voluntary and
non-coercive, even where differential abilities and endowments are present.
The voluntarism of labor supply decisions is the theoretical component that
allows for assessing forgone alternatives as costs of pursued action perceived
by the agent ex ante to be beneficial.
8. In addition, there have been several studies about workers conditions
and the use of monitoring programs against employee abuse in factories
(Varley 1998; Esbenshade 2004: 165198).
9. In addition, our findings can reject the strong claims often made by
activists that sweatshop workers are universally exploited.
10. We report the demographic and household characteristics in Appendix 1.
11. Voluntarily adopted codes of conduct, corporate policy, and
community-based consumption standards do not necessarily entail the same
welfare costs as government imposed standards (Powell and Zwolinski 2010).
12. Boettke, Coyne, and Leeson (2008) identify the difficulty of accomplishing successful institutional change. They argue that the stickiness of
institutional changes depends on the source of the institutional change (being
either foreign or indigenous) and the type of change (either bottom-up or
top-down oriented). Ultimately, they conclude that institutional changes that
are indigenously developed and based on bottom-up processes will be most
successful. Sweatshop employments based on voluntary participation fit these
criteria for effective and persistent institutions.

References
Aitken, B., A. Harrison, and R. Lipsey. (1996). Wages and Foreign Ownership:
A Comparative Study of Mexico, Venezuela, and the United States.
Journal of International Economics 40: 345371.

558

The American Journal of Economics and Sociology

Basu, K., and H. Zarghamee. (2009). Is Product Boycott a Good Idea for
Controlling Child Labor? A Theoretical Investigation. Journal of Development Economics 88: 217220.
Bellamy, C. (1997). State of the Worlds Children Report 1997. New York:
Oxford University Press.
Bhagwati, J. (2004). In Defense of Globalization. New York: Oxford University
Press.
Boettke, P., C. Coyne, and P. Leeson. (2008). Institutional Stickiness and the
New Development Economics. American Journal of Economics and
Sociology 67: 331358.
Brown, D., A. Deardorff, and R. Stern. (2003). The Effects of Multinational
Production on Wages and Working Conditions in Developing Countries. National Bureau of Economic Research, Working Paper No.
9669.
Campos, S. (2008). Ex-Adidas Factory Worker Describes Poverty Wages in El
Salvador. Maquila Solidarity Network, November 13.
Denzau, A. T., and D. C. North. (1994). Shared Mental Models: Ideology and
Institutions. Kyklos 41(1): 331.
Dugger, W. M. (1996). Redefining Economics: From Market Allocation to
Social Provisioning. In Political Economy for the 21st Century: Contemporary Views on the Trends of Economics. Ed. C. Whalen, pp. 3143.
Armonk: M. E. Sharpe, Inc.
Esbenshade, J. (2004). Monitoring Sweatshops: Workers, Consumers, and the
Global Apparel Industry. Philadelphia: Temple University Press.
Hall, J., and P. Leeson. (2007). Good for the Goose, Bad for the Gander:
International Labor Standards and Comparative Development. Journal of
Labor Research 28: 658676.
Harrison, A., and J. Scorse. (2010). Multinationals and Anti-Sweatshop Activism. American Economic Review 100: 247273.
Kristof, N. (1998). Asias Crisis Upsets Rising Effort To Confront Blight of
Sweatshops. The New York Times, June 15.
. (2009). Where Sweatshops are a Dream. The New York Times, January
15.
Lipsey, R., and F. Sjoholm. (2001). Foreign Direct Investment and Wages in
Indonesian Manufacturing. National Bureau of Economic Research,
Working Paper No. 8299.
Matthaei, J. (1984). Rethinking Scarcity: Neoclassicism, Neomalthusianism, and Neomarxism. Review of Radical Political Economics. 16:
8194.
Martin, N., and V. Storr. (2008). On Perverse Emergent Orders. Studies in
Emergent Orders 1: 7391.
Powell, B. (2006). In Reply to Sweatshop Sophistries. Human Rights Quarterly 28: 10311042.

Sweatshops, Opportunity Costs, and Compensation

559

Powell, B., and D. Skarbek. (2006). Sweatshop Wages and Third World
Workers: Are the Jobs Worth the Sweat? Journal of Labor Research 27:
263274.
Powell, B., and M. Zwolinski. (2010). The Economic and Ethical Case for
Sweatshops Reclaimed. Working Paper. Suffolk University.
Power, M. (2004). Social Provisioning as a Starting Point for Feminist Economics. Feminist Economics 10: 319.
Sachs, J. (2005). The End of Poverty: Economic Possibilities for Our Time. New
York: Penguin Books.
Sen, A. (1999). Development as Freedom. New York: Random House Inc.
Subrick, J. Robert, and Peter J. Boettke. (2002). Rule of Law, Development,
and Human Capabilities. Supreme Court Economic Review 10: 109126.
United Nations. (2010). UNESCO Institute for Statistics.
Varley, P. (ed). (1998). The Sweatshop Quandary: Corporate Responsibility on
the Global Frontier. Washington, DC: Investor Responsibility Research
Center.
Weiss, R. (1994). Learning from Strangers: The Art and Method of Qualitative
Interview Studies. New York: The Free Press.
Zwolinski, M. (2006). SweatshopsDefinitions, History, and Morality. In
Social Issues in America: An Encyclopedia. Ed. J. Climent. Armonk: M.E.
Sharpe.
. (2007). Sweatshops, Choice, and Exploitation. Business Ethics Quarterly 17: 689727.

Gender

Female
Male
Male
Female
Female
Male
Male
Female
Female
Female
Male
Male
Male
Male
Male
Female

Subject

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

36
24
21
43
25
25
19
24
31
36
28
25
24
21
30
21

Age

Y
P

N
Y
Y
Y
Y
N

N
Y
N

Spouse

9th Grade
High School
High School
High School
High School
High School
High School
High School
High School
7th Grade
High School
High School

High School
High School
9th Grade

Year
Completed

16
15

16
17

18

19
16
14
17

Age

Highest Level of
Education Attainment

1
1
0
1
0
2
0
**
2
3
2
0
0
0
2
0

Number

9
4

24

12, 1.5

7, 8
18, 15, 10
7, 5

9, 4

Age

Yes
No

No

Yes

No
Yes
Yes
Yes

Yes

Of School
Age*

Children

Interview Subject Demographics

Appendix

Yes

Yes

Yes

Yes
No
Yes

Yes

School
Attendance

Nuclear Family
With Partner
Extended Family
Nuclear Family
Parents In Home
Parents In Home
Nuclear Family
Nuclear Family

Nuclear Family
Parents In Home
Parents In Home
Parents In Home
Nuclear Family
Parents In Home

Household
Composition

560
The American Journal of Economics and Sociology

Female
Male
Female
Female
Female
Female
Female
Male
Male
Male
Female
Female
Male
Female
Female

17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

42
29
37
36
34
39
29
27
27
40
33
23
28
24
24

Age

Y
Y

Spouse

High School
High School
9th Grade
High School

6th Grade
9th Grade
High School

8th Grade
6th Grade
9th Grade
High School
8th Grade

Year
Completed

15
13
15

18

15

Age

3
1
3
1
4
0
0
2
2
2
3
2
0
0
2

Number

Yes
No
Yes
No
Yes

No
No

Yes
Yes

Yes

Of School
Age*

Children

17, 15, 12
3
16, 11, 9
21
14, 12, 12, 9

4, 2
6, 2

17, 16, 4
7, 2

8, 3

Age

Continued

Yes

Yes

Yes

Yes
Yes

Yes

School
Attendance

Nuclear Family
Parents In Home
Parents In Home
Parents In Home
Single Parent
Alone
Parents In Home
Parents In Home
Parents In Home
Nuclear Family
Single Parent
Nuclear Family
Parents In Home
Parents In Home
Nuclear Family

Household
Composition

*Of School Age (718); **Subject was three months pregnant at the time of interview; Child dropped out after 9th grade to learn to be an auto mechanic.

Gender

Subject

Highest Level of
Education Attainment

Appendix

Sweatshops, Opportunity Costs, and Compensation

561